          |
Meta-Analysis: Effects of Educational Technology on Student OutcomesGlossary of Research Termsaffective: Influenced by or resulting from human emotions. analysis of variance (ANOVA): An analysis of the variation in the outcomes of an experiment to assess the contribution of each variable to such variation. behavioral: Pertaining to how people behave or act. categorical variable: A variable that organizes respondents into groups of items based on classification by type or by kind (e.g., political affiliation, gender, ethnicity). causal inference: A determination that changes in one variable are related to changes in another variable. code: To prepare for analysis a given set of raw data or items by translation into a set of quantitative or qualitative symbols. cognitive: Pertaining to how people understand and acquire knowledge. conditioning variables: Demographic and other background information about a respondent that has an effect on the variation in the outcomes of the experiment. control group: A group whose characteristics closely match an experimental group in every aspect except for the experiment or intervention measures (e.g., a medicine or a lesson plan). The control group does not receive the experiment or intervention. correlation: A statistical measure of the degree of linear dependence between or among variables. It is expressed in the form of an index r that may vary from -1.00 to +1.00. Also called Pearson's correlation.
dependent variable: The variable of focus, or the variable whose value is predicted, measured, or acted upon by another variable or variables observed in the study, if there is any relationship with the variable(s) observed. effect size: A statistic used to measure the effectiveness of a treatment by measuring the distance between the means of a treatment group and a comparison group. empirical knowledge: Data and findings that are verified based on observation or experience. experimental research: Scientific investigation in which an investigator manipulates and controls one or more independent variables to determine their effects on the outcome (dependent) variable. F value: The measurement of distance between individual distributions. independent variable: The variable selected by the researcher whose value is known and whose value is systemically manipulated to determine the degree of its effect on the dependent variable. Also referred to as the treatment. mean: A measure that is calculated by dividing the sum of all values by the number of those values. Also called the arithmetic average. meta-analysis: The process or technique of synthesizing multiple research results by using various statistical methods to retrieve, select, and combine results from previous separate yet related studies to test a hypothesis. multicollinearity: A statistical phenomenon that occurs when two or more independent variables are so highly correlated that interpretation of the effects of their variation on the dependent variable is virtually impossible to determine. outlier: A value that lies outside of the general distribution. For example, in the data set "2, 78, 82, 84, 85, 89," the value "2" is an outlier. quantitative: Expressed or expressible as a measure of quantity or amount. quasi-experimental: A type of research design for conducting studies in field or real-life situations that allows researchers to make comparisons between the mean performances of groups that occur naturally as opposed to comparisons of group performances based on experimental or control-group designs. In this type of a design, the researcher may be able to manipulate some of the independent variables, but random assignment does not occur because the individuals studied must be studied as a part of the group to which they naturally belong. regression: Any of several statistical techniques concerned with estimating the mean value of one or more variables by knowing the value of at least one other variable. significance: The degree to which a research finding is meaningful or salient. The level of significance of a test of hypotheses is the probability of rejecting the null hypothesis when it is in fact true. standard deviation: A statistic that quantifies the dispersion of scores across a distribution by providing an average of the differences of all scores within the distribution from the mean. The more dispersed the scores are from the mean, the larger the standard deviation will be. t value: The results of a comparison between two group averages or means. threats to validity: Factors that can lead to flawed measurements of content or constructs within tests or evaluation instruments. variable: A characteristic that can vary or differ, such as age, location, or education level. variance: A measure of the spread or dispersion of scores within a distribution. A larger variance indicates that individual cases are further from the mean; a smaller variance indicates that individual scores are closer to the mean. windsorize: A way to edit data to adjust for outliers. Z score: A value that shows the location of a given score across a distribution and indicates how far the score is from the mean, as measured in standard deviation units. Sources American Heritage Dictionary of the English Language (4th ed.). (2000). Boston, MA: Houghton Mifflin. Ary, D., Jacobs, L. C., & Razavieh, A. (1996). Introduction to research in education (5th ed.). Orlando, FL: Harcourt Brace College. Brown, J. D. (1988). Understanding research in second language learning. Cambridge, England: Cambridge University Press. Hardy, D. (2002). Research methods in linguistics. Retrieved January 10, 2003, from http://www.engl.niu.edu/dhardy/courses/508/ Vogt, W. P. (1999). Dictionary of statistics and methodology: A nontechnical guide for the social sciences (2nd ed.). Thousand Oaks, CA: Sage. This glossary was compiled by Mary O'Kelly of NCREL's Resource Center and David Durian of NCREL's Center for Technology, December 2002.
|
|||
 |
 Contact Us | Privacy Policy |